1

Promoting Productive Employment in Sub-Saharan Africa. A Review

of the Literature1

Background paper prepared for the meeting of the Knowledge Platform Development

Policies in Accra, Ghana, 3-5 April, 2013

by

Adam Szirmai, Mulu Gebreeyesus, Francesca Guadagno and Bart Verspagen2

Version 9-04-2013

1 This paper was commissioned by the Social Development Department of the Ministry of Foreign Affairs of the

Netherlands. It serves as a background document for a shorter note on productive employment that summarises the

main findings of the longer paper. The background paper and the note build upon a draft note prepared by the secretariat

of the Platform Dvelopment Policies of the Ministry of Foreign Affairs of the Netherlands: Promoting Productive and

Sustainable Employment, revised draft 5 February 2013.

2 Contact: A. Szirmai, United Nations University- Maastricht Economic and Social Research Institute on Innovation and

Technology (UNU-MERIT), Keizer Karelplein 19, 6211 TC, Maastricht, The Netherlands, Tel. 31-43-3884469, email:

szirmai@merit.unu.edu

mailto:szirmai@merit.unu.edu

2

Table of Contents

1 Introduction ............................................................................................................................................... 4

2 The nature and size of the employment problem ....................................................................................... 6

3 Availability of data on (un)employment in Africa .................................................................................. 12

4 Causes and solutions to the slow growth of productive employment in Africa: review of existing

literature ........................................................................................................................................................... 16

4.1 Structural change ............................................................................................................................. 16

4.2 Skill mismatch ................................................................................................................................. 19

4.3 The role of SMEs ............................................................................................................................. 19

4.4 The role of innovation ..................................................................................................................... 20

4.5 Policies for productive employment ................................................................................................ 22

5 Emerging debates and contrasting views on how to promote productive and sustainable employment in

Africa ............................................................................................................................................................... 39

5.1 Finding African role models ............................................................................................................ 39

5.2 Agricultural led industrial development .......................................................................................... 39

5.3 Resource based industrialisation ..................................................................................................... 39

5.4 Non-traditional exports .................................................................................................................... 40

5.5 Creating employment in labour intensive modern agriculture. ....................................................... 40

5.6 Engines of growth and employment creation: Is manufacturing still important? ............................ 40

5.7 Role of foreign direct investment in employment creation ............................................................. 41

5.8 Promoting entrepreneurship in the informal sector ......................................................................... 41

5.9 Exploiting unlimited supplies of cheap labour ................................................................................ 42

5.10 Population policy ............................................................................................................................. 42

3

5.11 Is skill mismatch in Africa myth or reality? .................................................................................... 43

5.12 The nature and focus of industrial policy ........................................................................................ 43

References ....................................................................................................................................................... 45

4

1 Introduction This note provides a brief overview of current research and knowledge on employment trends and policies in

sub-Saharan Africa. The aim is to reflect on our present state of knowledge, identify gaps in our knowledge

and understanding, and contribute to evidence-based policy debates. The emphasis is on the productive and

sustainable nature of employment, rather than on the quantity of labour demand, or the rate of

unemployment. This is because of the specific situation in Sub-Saharan Africa (SSA), where the

employment problem does not primarily manifest itself as open unemployment, but as underemployment,

vulnerable and low quality employment. Open unemployment is scarce in Africa, but very large numbers of

the working population are employed in agricultural and the informal service sector where productivity and

earnings are low and there is a high degree of job vulnerability. In a development context, the term

“underemployment” refers situations where productivity and earnings are so low that a worker cannot make

a decent living in a normal work week and has to work very long hours to survive. Other connotations of

underemployment are that the job does not make use of the workers’ skills, education and experience.

Finally underemployment can also refer to situations where workers work less than they would like to work

(part-time work instead of full time work). “Vulnerability” refers to work with highly fluctuating and

uncertain returns, and without a stable and secure relation between employer and employee. Vulnerability is

an important aspect of unproductive labour. It is a typical characteristic of the informal sector.

Despite rapid growth in many sub-Saharan African countries over the past fifteen years, there is widespread

concern that this growth has not created sufficient productive employment to lift large numbers of the

population out of poverty (Kapsos, 2005; ILO, 2013; McKinsey, 2012; Fox and Sekkel Gaal, 2008). Access

to productive employment is essential for inclusion of the poor in society. Productive employment does not

only provide the poor with better incomes, it also stimulates learning and skills acquisition (World Bank,

2013). The insight that poverty reduction and social inclusion are linked to economic development via

improved job creation and productive employment represents an important shift in our thinking about socio-

economic development (see Kremer et al., 2009).

Economic growth may create productive employment by means of a combination of rapid growth of output,

innovation and upgrading, productivity increases and optimal utilization of abundant labour. Structural

change, i.e., shifts of employment between sectors, may promote productive employment by a shift towards

more dynamic and high productivity sectors that can absorb labour and provide jobs of better quality. In

present-day Africa, the production structure in many African economies is unbalanced, with an undue

reliance on exploitation of natural resources that cannot provide sufficient productive employment. There has

been insufficient structural change, among others as a result of premature deindustrialization (Tregenna,

2013).

5

Because the African employment problem is more one of quality of jobs rather than quantity of jobs, the

types of jobs that need to be created in Africa are jobs of sufficient quality, i.e. productive employment.

Following ILO (2009), we define productive employment as employment yielding sufficient returns to

labour to permit workers and their dependents a level of consumption above the poverty line. According to

this definition, whether a person is productively employed depends on the income from labour; the intra-

household dependency ratio (i.e. the number of people depending on the income); the labour income of other

employed members of the household, and other non-labour household incomes. The following indicators are

currently used to measure productive employment: 1) labour productivity growth (measured as annual

change in GDP per person employed); 2) employment-to-population rate (proportion of a country’s working-

age population that is employed); 3) proportion of the employed population living on less than US$1.25 a

day (the working poor); 4) the proportion of own-account and unremunerated workers (e.g., contributing

family workers) in the employed population (vulnerable workers).3 The last two indicators are newly defined

measures that capture job quality. Indicator 3) approximates how many people have jobs that cannot lift them

out of poverty, while indicator 4) indicates how many people work in precarious circumstances, because as

own account and family workers they are less likely to have a formal work arrangements ensuring continuity

of work and social protection.4

The term “decent work” completes the definition of “productive employment” by adding dimensions

referring to working conditions such as absence of coercion (no slavery, no child labour), equity at work

(equity of conditions and opportunities for all workers), security at work (health, pensions, security against

job loss), and dignity of work (Anker et al., 2002). Decent work also means decent working hours, i.e.

working not more than 48 hours per week (ILO, 2012).

The term “sustainable employment” is difficult to define independently of productive employment, and it

may not be needed as a separate category here. The term is often used to refer specifically to reduced job

vulnerability, but as such we understand it to be a part of productive employment, not as an alternative

concept.

Productive employment creation depends not only on changes in productive capacity and economic

structures, but also on supporting policies. Policies can provide incentives for better use of abundant labour

resources and enhance the productive capacity of the labour force through the development of human capital 3 Efforts to estimate trends in job creation in terms of employment quality and income distribution in the developing world has been made by the ILO, which is currently elaborating estimates of employment across economic classes

(ILO, 2013).

4 This definition of vulnerable employment might not be fully comprehensive because some wage workers can be worse

off than own account workers.

6

or policies supporting innovation and technological upgrading. Employment policies should be seen as part

of a much wider range of industrial policies, innovation policies, and economic policies promoting both

economic development and productive employment creation. However, policies that stimulate economic

growth and structural change may not create enough productive employment if society does not change in

terms of the institutions that underlie economic activity and employment relations. This includes formal rules

such as laws, but also informal mechanisms such as the nature of family relations, the mix between the

formal and the informal sector, and power relations between the economic elite and workers. These are

particularly relevant for the design of employment and population policy. Changes and transformations in

society may also play a large role in generating productive employment. Some of this comes under the

heading of “inclusive innovation”, which is a term that we use to describe technological, organizational, and

social innovation that lifts parts of the population out of poverty.

The rest of this study is organized as follows. Section 2 reviews the nature and size of employment problem

in SSA. Section 3 discusses the availability of data on employment and unemployment. Section 4 reviews

existing literature regarding the causes and the solutions to the slow growth of productive employment in

Africa. Section 5 presents emerging debates and contrasting views on how to promote productive and

sustainable employment in Africa.

2 The nature and size of the employment problem The unemployment rate in the SSA has been around 7.6% in the past 5 years (ILO, 2013), which seems to

suggest that only a small fraction of the working-age population is outside the labour market. Whether these

figures are sufficiently trustworthy is itself an interesting area of research, but, as noted already, we are not

only interested in open unemployment, but in particular in vulnerable employment, low quality employment

and underemployment.

With a growing workforce and not enough formal jobs created, job seekers resort to the informal sector. A

defining characteristic of the informal sector is that activities are non-registered. In consequence even when

informal enterprises employ wage labourers, these workers have no formal protection. In Africa, the informal

sector is mostly made up out of very small–scale non-agricultural activities, with employment characterized

by self-employment or employment in a family business.5 A large segment of the informal sector is involved

in the provision of a broad range of services such as barbering, repair, food service, street vending and other

trading activities, and telecoms, like mobile phone kiosks or cards (Fox and Sekkel Gaal, 2008). The

relevance of the informal sector in African economies is documented in several contributions (e.g., Sekwati

5 Even though agriculture shares several characteristics of the informal sector, the term informal sector as commonly

used refer to non-agricultural informal activities (e.g. informal manufacturing, informal services, informal construction).

7

and Narayana, 2011 and World Bank, 2011 for Botswana; Palmer, 2007 for Ghana; Luebker, 2008 for

Zimbabwe; Pollin, 2009 for Kenya; Kweka and Fox, 2011 for Tanzania).

Informal sector activities are present in both urban and rural areas, but are more widespread in urban areas.

But according to Haggblade et al. (2010), the rural nonfarm economy (RNFE) is large and expanding in

developing countries and income from rural nonfarm activities represents 35% of total income of the rural

African populations. Much of these activities are informal. For agricultural householders, the expansion of

rural non-farm activities stems from the necessity to diversify risk, counterbalance seasonal income swings,

and finance agricultural investments. Such expansion has meant that RNFE has started to be seen as a source

of income and employment, and so as a strategy towards poverty reduction (see Dimova and Sen, 2010, for

Tanzania; Stifel, 2008, for Madagascar; Bezu and Barrett, 2010, for Ethiopia). According to the empirical

analysis by Reardon (1997) and Barrett et al. (2001), in Africa non-farm rural income is positively associated

with households’ welfare, but entry and mobility barriers exist in the high-return niches of RNFE and greater

nonfarm income diversification yields higher growth in earnings and consumptions.

In the last decades Africa has experienced a shift away from agriculture to other sectors, mainly services, but

little expansion of manufacturing employment (see section 4.1). The service sector is more productive than

subsistence agriculture, but less productive than manufacturing. Despite this shift, agriculture continues to be

the largest source of employment in Africa. But agricultural employment remains highly vulnerable..

According to Ncube (2008), employment in the agricultural sector is characterized by long and irregular

working hours, lack of social benefits, job insecurity, contract and casual labour, and child labour. With

respect to the nature of employment in agriculture, small farms and subsistence agriculture dominate Africa’s

agriculture and only few countries (e.g. Burkina Faso) manage to raise the grain output of their small farms

(Anríquez and Bonomi, 2007; Wiggins, 2009; Aliber and Hart, 2009 and Baiphethi and Jacobs, 2009).6

Smallholders are heterogeneous with respect to access to resources –such as land- and markets and the

poorest farmers face high obstacles to move away from subsistence agriculture towards higher-value

horticultural and livestock products (Staatz and Dembele, 2007).

In the service sector, employment tends to take the form of self-employment or family businesses, rather

than wage employment. Thus, it is also characterized by high degrees of informality, and therefore high

degrees of job vulnerability. Wage employment, instead, is more likely in manufacturing, the employment

share of which has been shrinking in the last few decades. Adjustment policies in the 1990s have also

resulted in losses of formal jobs in the public sector which is another important source of formal wage

employment (Fox and Sekkel Gaal, 2008). The nature of structural change in SSA explains why, despite high

economic growth, vulnerable employment has not significantly decreased in SSA. The proportion of workers

6 For estimations on the extent of small farming, see Modrego et al. (2006).

8

in vulnerable employment decreased only marginally from 83% in 1991 to 82% in 2000 and 77% in 2012.

These are still very high rates and comparable only to South Asia (ILO, 2013; UNECA, 2005). Apart from

demography and the nature of structural change in SSA, the informal sector continues to prosper due to lack

of skills (discussed in section 4.2), and increased income coming from other sectors (Fox, 2011).

Estimations of the degree of underemployment (defined as working less than 40 hours per week) reveal that

underemployment is prevalent in agriculture (and so in rural areas) and among young people. It is less likely

in formal employment and larger firms, and decreases with education (e.g. Denu et al., 2005 for Ethiopia;

Sakey and Osei, 2006, for Ghana; Jones and Tarp, 2012 for Mozambique). At the other extreme of

underemployment, there are people that work excessive hours in order to survive. Excessive hours

characterize male employment especially in urban areas. One of the countries with the highest prevalence of

people working excessive hours is Tanzania, where 54.3% of the population was working excessive hours in

2005 (ILO, 2010, 2012).

Youth unemployment

The unemployment problem in Africa is characterized by high heterogeneity across countries, high youth

unemployment, and high disparities by gender and geography (rural versus urban areas), and level of

education (Page, 2012). More than two-thirds of the population of Sub-Saharan Africa was under 25 years of

age in 2010 and this percentage is expected to increase in the next decades. 60% of Africa’s unemployed are

young, and youth unemployment rates are double those of adults in most African countries.7 Even in

countries where the youth unemployment rate is relatively low, it is often more than twice as high as the

national average. A very high proportion of young people are poor: on average 72% of the youth population

in Africa has to live on less than $2 per day. Young people often work in the informal sector and are less

likely to be wage-employed or self-employed (World Bank, Africa Development Indicators 2008/2009).

While the average young worker in Africa is in family-based agriculture, other important occupations are

services and sales, and 13% are business owners (African Economic Outlook, 2012). In countries with high

youth unemployment, unemployment often goes hand in hand with low quality of jobs (vulnerable

employment) and lack of information about job seekers and job opportunities. In these contexts, skill

mismatch is often another aspect of the employment problem (Fox and Sekkel Gaal, 2008; Garcia and Fares,

2008; African Economic Outlook, 2012; World Bank report 2013).

7 In the literature, youth is defined as people aged between 15 and 24.

9

According to the African Economic Outlook (201), youth unemployment is particularly acute in middle-

income countries (MICs). 8 Youth unemployment is a problem because unemployed youth can get frustrated

and cause instability (as happened in North Africa) and because initial long-term unemployment negatively

effects lifetime earnings and future professional development. Country level data suggest that youth

employment is largely a problem of quality in low income countries (LICs) and one of quantity in middle

income countries (MICs). This has to do with economic growth and its structural implications: when

countries grow richer, they become more capital-intensive and demand higher quality goods. These two

forces generate a reduction in the demand for low-skilled labour (and relative increase in the demand of

high-skilled labour) and a shrinking of the informal sector (that produces low quality goods). So, in LICs,

young people work mainly in the informal sector, where wages are low, i.e. labour is of low quality. In

MICs, the informal sector is smaller and the formal sector is too small and demands high skills, so high-

skilled labour competes for too few jobs and low-skilled labour is left out of the labour market. This results

in high open youth unemployment.

Despite the fact that the informal sector suffers from low productivity and low wages, it still presents

opportunities and is part of the solution to the problem of Africa’s youth unemployment. The same applies to

the rural sector that has the potential to become an engine of inclusive growth and youth employment.

Farming, in fact, often branches out into household enterprises (Fox and Pimhidzai, 2011).

A large youth cohort can also yield opportunities, if growth is rapid and appropriate policies help to take

advantage of the demographic dividend resulting from having a larger share of the population at working-

age. In this regard, investments in human capital and policies to reduce the skill mismatch are essential

(Garcia and Fares, 2008; UNECA, 2011a; Africa Economic Outlook, 2012).

Differences in conditions

Policy debates on productive employment and employment creation should take differences of conditions

and opportunities in sub-Saharan Africa into account. Several classifications have been proposed for SSA.

The World Development Report 2013 categorizes countries according to urbanization, demography, natural

endowments, and strength of institutions. According to these dimensions, countries can be categorized into:

agrarian, urbanizing and formalizing countries; countries with high youth unemployment and aging societies;

resource-rich countries and small island countries; and conflict-affected countries. In agrarian economies,

people cannot afford to be unemployed and have to accept jobs with low earnings and underemployment.

8 According to current World Bank classifications, upper middle income countries (UMICs) in SSA include: Angola,

Botswna, Gabon, Mauritius, Namibia, and South Africa. Cameroon, the Republic of Congo, Cote d’Ivoire, Ghana,

Lesotho, Nigeria, Senegal, Swaziland, Tonga, and Zambia are lower middle-income countries (LMICS).

10

Therefore, wage employment is not representative of the working status of the majority of the population, so

the rates of underemployment and vulnerable employment are more relevant than unemployment. Small

island countries, like Mauritius, are constrained by the low economies of scale or specialization. The

Mauritian case shows that strategic policies and strong institutions may compensate for them. For this kind

of countries, it is needed to establish links with nearby economic centres, maximizing the benefits of

migration, and exploiting niche markets as possible way out. In countries with high youth unemployment,

unemployment is often coupled with low quality of jobs and lack of information on job seekers and

opportunities.

Based on empirical evidence from an analysis of household and labour force surveys in 16 African countries

(AfDB, 2012), Page (2012) classifies countries according to the degree of informality of their labour market

and level of GDP. Countries with well-structured labour markets and low levels of informality (Southern

cone) currently face high rates of unemployment; lower income countries with high degrees of informality,

like Ethiopia, Ghana, Tanzania and Uganda, present relatively lower unemployment; finally there are

countries with large informal sectors and high unemployment rates, like Kenya, Mali, Zambia and Zimbabwe

(Page, 2012). In South Africa, where labour market legislations do not guarantee employment opportunities

in the informal sector, the employment problem translates into high unemployment. In fact, the economic

performance of South Africa since 1994 has been rather disappointing, with a growing unemployment rate

estimated between 26% and 40%, if discouraged workers are included (Rodrik, 2006; Kington and Knight,

2004, 2007).

Moving to more policy-related classifications, Collier and O’Connell (2008) distinguish three categories of

countries: (i) High opportunity coastal, resource-scarce countries (ii) low opportunity land locked resource-

scare countries (iii) resource rich countries. UNECA (2011b) suggests to categorize countries according to

the geographical characteristics (resource endowments, landlocked, non landlocked) and demographic

characteristics (population size, density, age composition). In resource-abundant countries, sectoral policies

should favour their resource sectors. For landlocked countries, it is crucial to leverage on regional

integration. Countries with large population should relax rules on competition to allow domestic firms to

reap the benefits of economies of scale and thus prepare them for international competition.

Following Collier and O’Connell (2008), UNCTAD (2011) classifies countries by their level of

industrialization in 2010 and growth performance between 1990 and 2005. The report distinguishes (i)

forerunners, (ii) achievers (high level, low growth), (iii), catch-up countries, (iv) falling-behind countries, (v)

infant countries. Challenges and policy agendas vary across these groups. For the forerunners, policy

priorities focus on the shift towards industries with higher technological intensity and value addition (like

machinery and equipment or precision instruments) and creation of networks among existing firms.

Achievers’ strategies should be twofold: advancing technological capabilities and entrepreneurship in new

11

manufacturing sectors and upgrading in core existing industries to enter into high-margin segments of the

production chain. In catching-up countries, growth rates in certain industries are mainly dominated by few

large firms whose linkages amongst themselves and with the rest of the firms (small and informal) need to be

strengthened. In the medium-term labour intensive manufacturing can be a promising sector to target.

Falling-behind and infant-stage countries need to support entrepreneurship and acquisition of basic

managerial and technical competencies in order to move from natural resource extraction or agricultural

commodity production to a higher degree of processing.

In Figure 1, we take South Africa, Mozambique and Botswana as examples and show a possible way to

synthesizing these classifications. A dimension that is shared by several of these classifications is the

geographical one. In this respect, we account for resource endowment and whether a country is landlocked or

coastal, where a country is considered a natural resource economy if it generates more than 10% of GDP

from primary commodity rents (Collier and O’Connell, 2008). The second shared dimension is related to

demography. Based on Collier and O’Connell (2008), UNECA (2010) and the World Bank (2012), the size

of population is included. Connected to demography is the issue of youth unemployment, as evidenced by

the World Bank report 2013. Following the classifications by the World Bank (2013) and Page (2012),

labour market characteristics and strength of institutions are captured by the share of urban population and

share of informal sector in the economy. Finally, we account for the industrial and economic performance of

the country by looking at level of GDP, average GDP growth from 1960 to 2000 and the level of

industrialization (as measured by UNCTAD, 2011, Collier and O’Connell, 2008 and Page, 2012).

Figure 1. A synthesis of the classifications of SSA countries

Sources: Authors’ elaboration based on data collected from: for geographical variables and the average GDP growth from 1960 to 2000, Collier and O’Connell (2008); for population size and degree of urbanization, World Development

-200

20406080

100Coastal

Resourceendowment

Population size

Youth unemployment

Degree ofurbanizationDegree of informality

Level of GDP

Average GDP growth,1960-2000

Industrialperformance

Botswana

Mozambique

South Africa

12

Indicators (WDI) online; for youth unemployment and level of GDP, African Economic Outlook 2012 and WDI; for degree of informality, Ncube (2008); for industrial performance, UNCTAD (2011). Notes: Being a coastal or landlocked country is a dichotomous variable that takes the value of 1 if the country is coastal. Similarly, if a country is resource-rich according to the definition of Collier and O’Connell (2008), the variable assumes value of 1. Following UNCTAD (2011), countries have been ranked within the 5 categories, from forerunners to infant countries. Forerunners are assigned the highest value (5) and infant countries the lowest (1). GDP levels are measured as the GDP per capita (PPP valuation, USD) as in the African Economic Outlook (p. 240). Degree of informality is proxy by the contribution of the informal sector to GNI (Ncube, 2008, p. 5). All variables are taken as percentages with respect to the maximum values observed in SSA (Nigeria for the population size, South Africa for youth unemployment, Gabon for urbanization and level of GDP, Mozambique for informality, Botswana for GDP growth between 1960-2000).

This diagram allows comparisons across different SSA countries and so it can guide policy makers in the

selection of potential African role models. By comparing economic and industrial performance, and also

characteristics of the labour market –such as the degree of informality and the rate of youth unemployment-,

countries with similar pre-conditions can learn from each other. However, constructing this type of diagrams

for all SSA countries is not easy, due to lack of data especially concerning degree of informality and youth

unemployment.

3 Availability of data on (un)employment in Africa Data availability about employment is a serious constraint to research and analysis (DIAL, 2007; World

Bank, 2013). Data on employment normally derive from three main sources: labour force surveys,

production surveys (agricultural surveys, surveys of manufacturing, service sector surveys) and household

surveys. Labour force surveys provide most information about employment conditions, remuneration, hours

worked, labour market participation, and so forth. Production surveys have the advantage that output and

employment figures are from the same source, allowing for productivity analysis, but they do not provide

complete information about national employment trends and are usually restricted to larger enterprises.

Surveys of informal sector firms are held only very infrequently. Household surveys are important for

linking employment conditions to individual and household poverty, but usually they do not provide

sufficient detail on employment, its sectoral distribution and employment trends.

In many SSA countries labour statistics simply do not exist. Regularly repeated labour market surveys are

only held in Mauritius, South Africa, and Tanzania. In all other countries surveys are held irregularly,

sometimes with long intervals, so that it is almost impossible to chart trends in employment. Where data are

available there are important issues with regard to the statistical quality. Labour force surveys are often not

harmonized with industrial surveys. There is insufficient information about the nature of work in the

informal sector, especially with regard to underemployment and youth employment. There is an increasing

wealth of micro-datasets for Africa (e.g. van Biesebroeck, 2005; Rankin et al., 2006; Söderbom et al., 2006;

Arnold et al., 2008; Baptist and Teal, 2008; Amin, 2009; Shiferaw and Bedi, 2009; Sonobe et al., 2009;

Goedhuys and Sleuwaegen, 2010), but it is not clear to what extent the micro-data samples are representative

13

of the national populations. Overall statistical capabilities have been declining rather than improving, so that

data collection tends to depend on incidental donor support. Challenges for labour statistics are: data gaps,

because in some countries labour statistics do not exist or are not collected systematically; data quality

issues, when data are collected, there are often problems in the statistical production chain (use of

inappropriate definitions, questionnaire design, sampling frame, data entry and coding, etc.); planning,

coordination, and communication issues arise when different institutions collect and disseminate the data

(World Bank, 2013; Kingdon and Knight, 2007 for the case of South Africa). Table 1 summarises the

availability of labour force and employment statistics in SSA.

Table 1. Sources of data for productive employment in SSA

SSA Country Type of data available Coverage Periodicity of

data collection

Years of data

availability

Angola - - - -

Benin Integrated Modular Survey on living conditions

of households Whole country Every 2 years 1984/5,2006/7

Botswana

Labour Force Survey Whole country Every 10 years 1984/5, 2005/6

Living Standards Survey Whole country Every 7 years 1985, 1993,

2003, 2010

Informal Sector Survey Whole country No indication 2007

Burkina Faso Annual Survey on Household Living Conditions

(QUIBB) Whole country Yearly 1995, 2005, 2007

Burundi Survey 1-2-3 Bujumbura Irregularly 2005

Household Living standard Survey No indication No indication 1998

Cameroon Enquête Emploi Secteur Informel No indication No indication 1995, 2005

Living Standards Survey Whole country No indication 1995, 2005

Central

African R.

OECD/Eurostat No indication No indication 1995

Living Standards Survey Whole country No indication 2005

Chad Living Standards Survey Whole country No indication 2005

Congo, D. R. Survey 1-2-3 Whole country Irregularly 2004/5

Living Standards Survey Whole country No indication 2005

Congo

Enquête sur l'Emploi et le Secteur Informel

(ECOM)

Brazzaville and

Pointe Noire Irregularly 2009

Living Standards Survey Whole country No indication 2005

Côte d'Ivoire

Enquête sur la Situation de l'Emploi à Abidjan Abidjan Irregularly 2008

Household Living Standard Survey No indication No indication 1985, 1986,

1987, 1988,

14

1992, 1995, 1998

Eritrea - - - -

Ethiopia

Labour Force Survey

Whole country

excluding some

areas

Irregularly 1999, 2004/5

Living Standards Survey Whole country Every 5 years 1995, 2000,

2005, 2011/2

Ethiopian Rural Household Surveys (ERHS) Some rural areas Irregularly

1989, 1994,

1995, 1997,

1999, 2004, 2009

Gabon Enquête Nationale sur Emploi et Chômage No indication No indication 1993, 2011

Living Standards Survey Whole country No indication 2005

Gambia Integrated Household Survey Whole country Irregularly 2003/4

Ghana Living Standards Survey Whole country Irregularly

1987, 1988,

1991, 1998/9,

2005/6

Guinea Living Standards Survey Whole country No indication 1995

Guinea-Bissau - - - -

Kenya Integrated Household Budget Survey Whole country Every 10 years 1998/9, 2005/6

Lesotho - - - -

Madagascar Enquête Périodique Auprès des Ménages Whole country Irregularly

1993,1995,1997,

1999,2001, 2005,

2010

Malawi Integrated Household Survey Whole country Irregularly

1991, 1997/8,

2002, 2004/5,

2010/11

Mali Enquête Permanente Auprès des Ménages

(EPAM) Whole country Every 2 years

1995, 2004,

2007, 2010

Mauritius

Continuous Multi Purpose Household Survey

(CMPHS) Whole country Quarterly 1999-2012

Small and Large Establishment No indication No indication 2002 and 2007

Mozambique Integrated Labour Force Survey

Whole country,

excluding 4 districts Irregularly 2004/5

Living Standards Survey Whole country No indication 1995,2005, 2010

Namibia Labour Force Survey Whole country Irregularly 1997, 2000, 2008

Living Standards Survey Whole country No indication 1995

Niger Living Standards Survey Whole country No indication 1995

15

Nigeria

Labour Force Survey Whole country Quarterly 2009

Living Standards Survey Whole country Every 5 years 1986, 1992,

1997,2003

Rwanda Living Standards Survey Whole country No indication 1995, 2005

Senegal Enquête de Suivi de la Pauvreté (ESPS) Whole country Irregularly 2005/6 and 2011

Living Standards Survey Whole country No indication 1995, 2005

Sierra Leone Integrated Household Survey Whole country Irregularly 2003-2004

Somalia - - - -

South Africa Labour Force Survey Whole country Quarterly 2000-2012

South Sudan - - - -

Sudan Household Living standard Survey No indication No indication 1988

Swaziland Household Living standard Survey No indication No indication 1985, 1995

Tanzania

Integrated Labour Force Survey Whole country Every 5 years 1995, 2000/1,

2006, 2010/11

Household Living standard Survey No indication No indication 1991,2000, 2001

Kagera Health and Development Survey Kagera Region No indication 1991, 1992,1993,

1994, 2004, 2010

National Panel Survey Whole country No indication 2008, 2010

Togo Living Standards Survey Whole country No indication 2005

Uganda Urban Labour Force Survey

Main

cities/metropolitan

areas/regions

Yearly 2002, 2009

Living Standards Survey Whole country No indication 2005/6, 2010

Zambia Labour Force Survey Whole country Irregularly 1986, 2005

Living Standards Survey Whole country No indication 2005

Zimbabwe Labour Force Survey Whole country Irregularly 1993, 2004

Household Living standard Survey No indication No indication 1990,1993, 1995

Source: Authors’ elaboration, based on national statistical offices, ILO9, UNECA/AfDB10 and World Bank (2012), table 9. Notes: This table does not include production surveys and production censuses. It also excludes SSA countries with populations of less than 1 million people.

The documented low frequency of data collection and low comparability of labour statistics hampers the

development of labour market information analysis (LMIA) systems. According to Sparreboom and Albee

(2011), "the state of LMIA systems in sub-Saharan Africa is an important reason why many countries fail to

9 http://laborsta.ilo.org/applv8/data/SSM3_NEW/E/SSM3.html#A 10 http://ecastats.uneca.org/acsweb/rrsf/en-us/baselineinformation/datadevelopment.aspx

16

formulate proactive employment and labour policies. Such policies, including ambitious but realistic targets

that are consistently monitored and evaluated, require effective LMIA systems based on regular data

collection and analysis. Strengthening LMIA systems and improving the availability of labour market

indicators is therefore essential to ensure better labour market outcomes" (ibid., p. 5).

4 Causes and solutions to the slow growth of productive employment in Africa: review of existing literature

4.1 Structural change There is a strong correlation between high shares of agriculture in GDP and low levels of per capita GDP.

The implication is that in poor countries agriculture may contribute substantially to employment, but this is

often low quality employment due to low productivity in traditional agriculture. As agricultural productivity

increases, the share of agriculture in GDP and employment will decline. The redundant workers in

agriculture will have to be absorbed in other sectors through a process of structural change. It is important to

note that the necessity of structural change should not lead to a neglect of African agriculture, as was

practiced in the period 1950-1980. Making agriculture more dynamic is an essential element of the process

of structure change and should figure prominently in economic policy making.

Figure 2. Agriculture as % of GDP

Source: World Development Indicators online.

Sectors and activities that can potentially absorb workers leaving traditional agriculture include commercial

farming and production of labour intensive higher value added crops; the rural and urban informal service

sector; the formal service sector, in particular business services, tourism, transport, logistics and distribution;

mining; construction; manufacturing and the public sector. These sectors differ greatly in terms of their

R² = 0,7316

0

10

20

30

40

50

60

373

1.0

49 1

.512

1.8

09 2

.399

3.4

12 4

.636

5.5

82 6

.779

8.6

69 1

0.23

4 1

3.09

9 1

5.07

8 1

7.31

0 2

1.26

1 2

6.20

8 3

5.24

6 4

0.37

0 4

8.11

2

Agriculture as % of GDP

ag %

17

opportunities to generate productive employment. Manufacturing and business services typically provide

productive jobs, while informal services and traditional agriculture provide jobs of less quality (McKinsey,

2012).

Today, apart from Government and social services, stable employment (as opposed to vulnerable

employment) in Africa is mostly concentrated in the manufacturing, construction, transport and

communication, and finance and business services (McKinsey, 2012). The observed size distribution of the

manufacturing firms in the developing countries is bi-modal, one mode at the small size group and another at

the large one. This is known in the literature as ‘industrial dualism’ (Tybout, 2000). The few large formal

firms provide products to niche or protected markets, while the many small low-productivity firms at the

bottom size distribution provide low-quality products to the domestic market. The latter generates low-

paying jobs and few productive employment opportunities (Dihn et al., 2012).

The past experiences with African manufacturing have been disappointing (e.g. Szirmai and Lapperre 2001

for the case of Tanzania). Since the mid 1980s, many countries in Africa have been experiencing de-

industrialization and the contribution of manufacturing to employment creation has been rather limited, with

other sectors tacking up the slack. Because agriculture and services are the sectors that are more

characterized by informality, the jobs that this type of structural change has generated are generally

vulnerable or characterized by underemployment, hence low quality jobs (among the others, McMillan and

Rodrik, 2011; McKinsey, 2012; Dihn et al., 2012; Page, 2013). McMillan and Rodrik (2011) show that,

despite a regional trend of growth-reducing structural change, SSA countries present high heterogeneity,

with Zambia and Nigeria experiencing structural change towards agriculture and Ethiopia, Ghana and

Malawi, instead, experiencing growth-enhancing structural change (towards manufacturing). Rodrik (2006)

argues that high wages and rigidities of the labour market, resulting from the strong position of trade unions,

are only the proximate cause of unemployment in South Africa. The process of structural change away from

the non-mineral tradable sector and the weakness of export-oriented manufacturing are the deeper causes of

relatively low growth and high unemployment. According to Hausmann (2007), the binding constraints to

growth in the tradable sector are: 1) level and volatility of the real exchange rate; 2) trade policy (and in

particular, tariff protection in intermediate goods); 3) the logistics system and high input costs resulting from

limited competition, 4) labour market constraints (high labour costs, rising wage differential between union

and non-union workers and skill mismatch); 5) obstacles to structural transformation, linked to specialization

in mining and low capabilities in other industries; 6) rules concerning the Black Economic Empowerment

(BEE).

Extractive industries (mining) present few employment opportunities and weak forward and backward

linkages to the rest of the economy. Diversification of the production and export structure, and mechanisms

to channel the wealth generated by resource extraction in the rest of the economy are crucial for how an

18

economy benefits from natural resources. The 2013 World Development report presents Norway and Papua

New Guinea as cases of successful management of natural resources revenues for diversification.

Too little is known about the role of the construction sector in structural change and employment creation,

even though it is an important sector in terms of the quantity of labour it employs. In Africa the construction

section creates both formal and informal employment.

The public sector is a source of formal employment in the service sector, but budgetary constraints and more

critical views of the potential of the public sector impose limits on public sector job creation.

The informal urban and rural service sector employs a large proportion of workers in SSA. As argued above,

this is often vulnerable and low quality employment. Data on earnings for self-employed persons and family

workers are hard to find (Fox and Gaal, 2008). The scarce evidence shows lower earnings than in the formal

sector (some data available in labour surveys of Uganda in 2001; Ghana in 1998; Senegal in 2001).

However, in rapidly growing economies, the informal sector earnings also tend to grow. Moreover, earnings

in the informal sectors are still higher than those in the agricultural sector. These are some of the reasons

why a solution to poverty in Africa should include the informal sector (Fox and Gaal, 2008; Sparks and

Barnett, 2010). One should realize that the informal sector is highly heterogeneous (Grimm et al. 2012).

Recent research argues that policy attention should focus on the most dynamic entrepreneurs and firms in the

informal sector which have the potential of rapidly expanding employment (e.g. Sonne, 2011)..

A way to tackle the issue of job creation in Africa is by investing in agro-industry and in labour-intensive

manufacturing and services. A shift of light manufacturing activities from East and South East Asia to Africa

is conceivable, given the labour cost advantage and the abundance of raw materials but it requires

investments in human capital and improvement of the business environment (Harrison et al., 2011; Clarke,

2011; Dihn et al., 2012; Leipziger and Yusuf, 2012; Page, 2012; McKinsey, 2012). In order to expand job

creation, Africa should invest in agriculture by expanding commercial farming and shifting production

toward more labour-intensive, higher-value-added crops; consider manufacturing as an engine of job creation

and, in particular, labour-intensive light manufacturing in pre-transition and transition countries, agro-

processing in countries with large agricultural sectors higher value-added exports in diversified countries,

and manufacturing for domestic markets in oil exporters (McKinsey, 2012).

In all these contributions, however, the focus is more on job creation as such, rather than the creation of

productive employment. Lavopa and Szirmai (2012) review the literature on the link between

industrialization, employment and poverty reduction. They discuss evidence that manufacturing contributes

significantly to growth, and via growth to employment creation and poverty reduction and that

manufacturing jobs tend to be of high quality in the sense that they pay higher wages and offer more indirect

benefits. An exception to the focus on job creation for African studies is Dihn et al. (2012), who argue that:

19

“the ongoing redistribution of cost advantages in labour-intensive manufacturing presents an opportunity for

Sub-Saharan Africa to start producing many light manufactures, enhance private investment, and create

millions of productive jobs” (ibid. p. 4).

4.2 Skill mismatch African countries have been extremely successful in expanding their education systems since 1950. They

have invested heavily in education at all levels and enrolments and graduations have increased dramatically

(Szirmai, 2013, chapter 7, Barro and Lee, 2010). Nevertheless, this has not translated into acceleration of

growth, structural change and catch up in Africa. The modern debate on the role education asks why this is

the case.

A very brief summary of the strands in this debate is as follows:

1 Investment in education affects economic performance with very long delays (of up to decades) and is

also dependent on complementary factors such as inflow of capital and knowledge which challenges the

acquired skills. In the 1950s, Africa had a huge skill gap with the rest of the developing world. Sixty

years later, it is better placed to profit from its accumulated stock of human capital.

2 In contrast to the optimistic analysis under point 1, recent research suggests that quantitative advance in

enrolment and graduation hides large skill gaps. The focus in education policy should be on improving

cognitive skills (Hanushek and Woessman, 2007, 2008).

3 There is a skills mismatch between what is being learned in educational institutions and what is required

by the labour market (World Bank, 2013; African Outlook 2012). The skills mismatch involves

insufficient attention for professional, agricultural, vocational and middle level technical training,

insufficient attention to on-the-job training and overschooling resulting in brain drain. But there is a

debate whether the mismatch is caused by shortcomings in the educational system or by distorted

financial and institutional incentives (Dihn et al., 2012; World Bank, 2013; Sekwati and Narayana, 2011;

Okunola et al., 2010).

With regard to the last point, light manufacturing activities like sewing in the apparel sector, require so

modest skills that industrialization could easily be ignited from the expansion of these sectors (and some

manufacturing even require the similar skills to those of the agricultural sector). Vocational training could be

offered by the State via public-private partnerships, starting from large firms in formal and informal

industrial clusters (Dihn et al., 2012).

4.3 The role of SMEs Developing countries are generally characterized by dualism at different levels of the economic and social

structure. Duality also manifests in industrial markets, made up of few large formal firms and a myriad of

small and mostly informal firms. Because job creation is mainly constrained by a lack of supply of jobs and

20

because the African private sector employment is dominated by small and micro firms, it is important that

policy addresses the issue of firm growth. There are few studies on this issue (e.g. Goedhuys and

Sleuwaegen, 2002; Bigsten and Gebreeyesus, 2007; Shiferaw and Bedi, 2009).

An analysis of the role of SMEs and entrepreneurship is relevant to this study for two main reasons. The first

is that SMEs and entrepreneurial activities (a great bulk of the informal sector) dominate the African

economy. The second is that if these micro firms are driven by opportunities and prove to be dynamic and

innovative, the constraints to their growth should be eliminated. In this way, more jobs could be created and

with the emergence of larger firms, informality and vulnerability could be greatly reduced (African

Economic Outlook, 2012).

Grimm et al. (2012) analyze a sample of informal entrepreneurs in seven capital cities of francophone West

Africa and disentangle the characteristics of the firms that have the potential to grow but did not yet do so

(constrained gazelles). They find that top performance is correlated with family wealth, which might imply

that access to credit is a binding constraint for dynamic SMEs. Although survivalist firms and constrained

gazelles have similar levels of capital stock, constrained gazelles show much better management skills (work

more hours than survivalists, keep more often books, have a much higher financial literacy) and seem to be

more entrepreneurial (react better to demand shocks and search actively for new clients) than survivalists.

Because returns to capital are also highest among constrained gazelles, they conclude that constrained

gazelles really represent an untapped growth potential (see also Sonne 2011).

Supporting SMEs can also take the form of supporting youth entrepreneurship. In order for this program to

be effective, however, they should be well-targeted and comprehensive. While many countries have put in

place programmes that cover job creation, training, and information services, only the Moroccan program is

positively evaluated. These types of programs need to build on reliable data on employment in Africa and

systematic processes of policy evaluation (African Economic Outlook, 2012).

4.4 The role of innovation The creation of increasing numbers of productive jobs is deeply entwined with a continuous process of

innovation. Innovation results in the upgrading of existing production and jobs, but also shifts to new

products and activities in the same sector or in different sectors (structural change). In low-income countries

innovation usually does not take place at the frontiers of international knowledge. It often takes the form of

adoption of internationally available technologies (e.g., Fu et al., 2011; Robson et al., 2009 for Ghana; Ola-

David and Oyelaran-Oyeyinka, 2012 for Kenya and Nigeria). But such technology acquisition is never

merely a process of passive imitation. It involves a highly creative process of selection, learning, adaptation,

upgrading and sometimes leapfrogging. The capacity to tap into global technology and knowledge flows

depends to a great degree on the development of capabilities and absorptive capacities. There is a large and

21

important literature on capability building and absorptive capacity, which is of considerable relevance for

sub-Saharan Africa (Abramovitz, 1986; Biggs et al. 1995; Cimoli et al. 2009; Cohen and Levinthal, Lall,

1987, 1990, 1992, 1994, 1996, 1998, 2000). Capabilities are categorized in many different ways. An

important distinction is that between production capabilities (the capability to operate a given technology),

adaptation technologies (the ability to adjust technology to new circumstances and conditions) and

innovation capability (the ability to start developing new technologies or upgrade existing ones).

Innovation depends not only on human capabilities but also infrastructural investment (e.g., Calderon and

Serven 2010; Ncube, 2010), for instance in ICT infrastructure. In recent years, rapid progress has been made

in Africa in creating ICT infrastructures, both using fibre technologies and satellite technologies (e.g. Special

Issue on “ICTs and Economic Transformation in Africa”, African Journal of Science, Technology,

Innovation, and Development, 2011; Mupela, 2011; Williams et al., 2011; Birba and Diagne, 2012), but

major obstacles still remain especially in thinly populated rural areas. The expansion of mobile telephony in

Africa is proceeding at an unprecedented rate, offering a host of innovative new opportunities.

One exciting new field of research links the literatures of entrepreneurship and innovation in the context of

developing economies. This research enquires into the conditions under which small and large entrepreneurs

can become more innovative and how policies could support this (see Gebreeyesus 2011, and Szirmai,

Naudé and Goedhuys, 2011 for a recent overview). The work of Haussmann and Rodrik (2003) on economic

development as self-discovery also focuses on the incentives for entrepreneurs in developing economies to

branch out into new activities (structural change as innovation).

In recent years there is increasing attention for the concepts of inclusive or pro-poor innovation – types of

innovation that contribute in important ways to poverty reduction and the needs of the poor. One strand of

research is that of inclusive innovation, or innovation at the bottom of the pyramid (Prahalad, 2006, Ramani

et al., 2012; for the African context, Ismail and Masinge, 2011), which focuses on the development of new

products that serve the needs of billions of poor people ‘at the bottom of the pyramid’. Goedhuys (2007)

discusses how in Tanzania foreign multinationals firms need to engage in collaborations with local partner in

order to better understand and meet the needs of the local poor. A second strand of research focuses

primarily on innovative entrepreneurial activities that create good quality jobs for poor people (Sonne, 2011).

Sonne (2011) studied the Indian landscape of small innovative pro-poor entrepreneurial firms and their poor

access to finance. The Indian financial system has evolved towards a dual system with traditional finance

providers, the banks, and a bunch of alternative financial institutions, specialized in supporting pro-poor

entrepreneur-based innovation. Access to finance for these types of firms is socially and economically

important: through access to finance, improved goods and services, increasing incomes and job opportunities

become available to rural communities. Sonne criticise the conventional micro-credit approaches and calls

22

for financial instruments to support a subset of somewhat larger dynamic and innovative firms (see also the

discussion of constrained gazelles above).

4.5 Policies for productive employment There is a lively debate about the nature of industrial policy and how industrial and innovation policies can

contribute to structural change, technological upgrading and the generation of productive employment (for

an overview see Naude and Szirmai 2012). Two interesting positions in this debate are provided by

Hausmann and Rodrik (2003) and Lin and Monga (2011). Hausmann and Rodrik interpret structural change

as a process of self-discovery, in which innovative firms discover where a country has a competitive edge.

Policy should aim at supporting such firms, because they bear more risks and costs than subsequent

followers who can imitate the leaders. Lin and Monga (2011) argue that a country can identify its latent

comparative advantage through comparison of its sector structure with similar countries at higher stages of

development. According to their framework, in the first step of an industrialization strategy, country should

identify the sectors in which they have latent comparative advantage. In order to do so, countries can look at

the list of tradable goods and sectors, produced in the last twenty years in growing countries with similar

resource endowments and with a per capita income about 100% higher than their own. Among these

industries, countries should favour industries where some domestic firms have already entered the market. If

domestic firms are not present in these industries, the government can attract FDI from world industry

leaders (by leveraging on lower labour costs or by creating EPZs and industrial parks, or by offering

temporary financial incentives).

A more statist position is taken by authors such as Ha-Joon Chang (e.g. Lin and Chang 2009; Chang 2012)

and Alice Amsden (2011), who argue that governments should take the lead in structural change by defying

static comparative advantage and ‘getting prices wrong’. But other authors argue that selective state

interventions require very high state capabilities, which are lacking in many sub-Saharan African countries.

Thus Tilman Altenburg argues that the neo-patrimonial state can be a serious obstacle to the effective

implementation of industrial policies in Africa (Altenburg, 2013).

Common trends in industrial policy in Africa include: attraction of FDI (especially for export-oriented

sectors); promoting of export-oriented industries; selective tariff protection and export taxes to create

incentives for local processing of raw materials; privatization of manufacturing public firms; sectoral policies

focusing on existing resources and light manufacturing (Marti and Ssenkubuge, 2009).

In 2003, ILO adopted the Global Employment Agenda which set forth several elements for developing a

global strategy for employment. At the core of this agenda there are economic development, social justice,

and the concepts of productive employment and decent work, as defined and discussed in Section 1. The

policies that allow these achievements are classified into policies addressing the economic environment

23

(trade, investment, innovation, policies for sustainable development and macroeconomic policies for growth

and employment) and policies that directly affect the labour market (policies for entrepreneurship,

employability by improvement of knowledge and skills, active labour market policies, social protection, and

productive employment). African policy makers are increasingly designing strategies that go in this

direction. These efforts are often undertaken together with the ILO and the African Development Bank Table

2 provides a summary view of the implementation of four categories of policies in sub-Saharan Africa: trade

policies, sectoral policies, innovation policies, employment policies. It is worth mentioning that, because

some of the sources of this information are official policy documents, it is difficult to evaluate the progress

of these reforms or their effective implementation. In table 3, we specify targeted sectors. In table 4, we

summarise existing literature on implementation and result of each of these policies. The table is also useful

to understand which policies have attracted most attention from academics and international organizations

and which areas have been neglected and are in need of further investigation.

Trade policy

Export promotion is a recurrent policy recommendation, especially for resource-abundant countries whose

production structure needs to be shifted towards sectors with more employment opportunities (World Bank,

2013). Currently, various African countries have in place export promotion agencies in charge of promoting

export-oriented manufacturing and services, like the BEDIA in Botswana. These however do not always

work properly (e.g., Lederman et al., 2010). Apart from export promotion agencies, governments set up a

wide series of financial and fiscal incentives for exporting firms (e.g., Belloc and di Maio, 2011). Export

promotion can also be achieved via Special Economic Zones (SEZs). SEZs attract investments that would

have not come to a particular country otherwise and therefore create additional jobs (Kingombe and te Velde,

2012). In Africa, export promotion is often mentioned in combination with attraction of FDI (Marti and

Ssenkubuge, 2009). FDI can be used strategically: as a diversification tool, and thus as a way to offer new

employment opportunities (Lin and Monga, 2011), as a way to increase exports (Hailu, 2010), and as a way

to absorb new knowledge and technologies and tap into global networks (discussed later). Exports can

contribute to technological upgrading through standards and quality controls. Standards and quality controls

are very important to compete in the international market. In order to stay competitive in the agricultural

sector, African countries have to continuously upgrade their position in global value chains (UNECA, 2008;

Page, 2013; Goedhuys et al., 2006; Iizuka and Gebreeyesus, 2012). There is an extensive debate on the

effects of selective trade protection and how it can impose market distortions and create inefficiencies. In the

African context, some studies are investing the extent with which tariffs and non tariff measures are still used

to protect domestic markets (e.g. Hoekman and Nicita, 2011; Jones et al., 2011). They show that trade

restrictiveness is still significant in low income countries, but that there is limited evidence of political

economy influences on the cross sector pattern of tariffs and reforms.

24

Sectoral policies

Sectoral industrial policies comprise all measures that target specific sectors over others (e.g. the automobile

sector in South Africa). According to UNECA (2011b), relevant tools of sectoral policies in Africa are

preferential credit, competition policy and public procurement favouring local manufacturers. Credit, in fact,

access to credit is a major obstacle to investment in the region (Ramachandran et al., 2009). Preferential

credit can be used to support firms, and especially SMEs, in priority sectors. Successful cases are Ivory

Coast (cement), Zimbabwe (wood products), South Africa (fertilizers), Ethiopia (flower) and Mozambique

(aluminum). According to UNCTAD (2011), competition policies are particularly important in the raw-

material sectors, which are often very important in the African context. In the last ten years, new regulatory

environments have spurred private investments in the sector (UNIDO, 2009). Very little empirical literature

is available in this policy area.

Innovation policy

As far as innovation policies are concerned, the set of policy instruments government can choose from is

generally quite ample (e.g. for South Africa: Lorentzen, 2009; Angola: UNCTAD (2008); Rwanda: Murenzi

and Huges, 2006; Mauritius: Wignaraja, 2002). Instruments like tax incentives and R&D subsidies are

offered only by South Africa and Rwanda (the latter has recently emerged as an innovation hub). By the

same token, few countries have engaged in regional policies (e.g., Ethiopia: Egziabher, 2000). According to

Aryeetey and Moyo (2012), priority in African industrial policy should be given to the adoption of new

technologies through R&D incentives, clusters and SEZs, and a strategic use of FDI by which foreign

investment would be channelled towards specific target sectors and incentives for the creation of linkages

with the rest of the domestic economy are in place. Marti and Ssenkubuge (2009) suggest using FDI more

selectively, by focusing on non-traditional activities (e.g. Botswana and Mauritius) and technologically

advanced activities (e.g. Ghana and Mauritius) and aiming at upgrading, technology and knowledge transfer

and entry into new value chains. Empirical evidence for African countries largely concentrates on FDI

attraction for technology transfer (e.g. Portelli, 2006). A rapidly expanding area of research focuses not so

much on specific instruments of innovation policy, but improving the function of national, regional and

sectoral innovation systems (Lundvall et. al, 2009). A key recommendation from this literature is to improve

the linkages and knowledge flows between economic actors, education institutions, research institutions and

governments.

Employment policy

The tools of employment policies are selected based on the Global Employment Agenda of the ILO. Policy

intervention should include government funded active labour market programs, aimed at employment in the

25

public and private sector; training and education incentives; job search assistance and other information

services; education systems reforms to tackle the skill mismatch and reforms of labour legislation (Page,

2012). Public works, i.e. subsidized temporary employment or labour intensive mega-projects, are often used

in African countries to create (temporary) employment. These often are large infrastructure projects (see

McCord and Slater, 2009). Public works’ programmes are often coupled with food-for-work programmes

(FFW). In FFW programmes, people are employed for public works and are paid in food. These programmes

guarantee at least the minimum essential quantity of food necessary to maintain good nutrition. Providing

food instead of money has several advantages, as Barrett et al. (2002) discuss. Few empirical studies evaluate

the effects of public works and active labour market policies on employment creation and its quality. One of

them is Rijkers et al. (2010). Recognizing the role of the informal sector, employment policies should

specifically be able to reach these firms and support their move into the formal sector (Marti and

Ssenkubuge, 2009). This applies to training as well (see the experience of Western African countries like

Ghana).

Population Policy

Population policy refers to policies that target demographic aspects such as fertility, mortality, population

growth, migration, distribution of people. The high rates of population growth prevalent in SSA cause

several problems, among which the problem of increasing youth unemployment discussed above. According

to the UN report World Population Policies (2009), 25% of African Governments started to adopt population

policy already in the mid-1970s and over time more and more countries followed. Today the majority of the

African Governments sees the level of fertility of their country as too high.11 Population policy in SSA

essentially aimed at providing information about, and access to, contraceptives. In some instances, these

programs yielded positive results in terms of fertility reduction (see Blacker et al., 2005 for a comparison

between Kenya and Uganda). According to Bongaarts and Sinding (2011), given that a good population

policy can decrease fertility by 1.0 to 1.5 births, prompt adoption of such programs in SSA could reduce

population by considerably more than a quarter-billion by 2050. While many SSA countries see migration as

a way to control unemployment and boosts their revenues via remittances, the negative impact of brain drain

on technological and socioeconomic development is of increasing concern (Adepoju, 2008). Migration and

brain circulation in SSA is primarily intra-regional. Due to poverty and deteriorating living conditions,

migration has increased in the last decades and international migration is also expected to rise as a

consequence of high (educated) youth unemployment. In 2009, 75% of the African Governments (75% of

the total) pursued policies to reduce the internal flow of migrants from rural to urban areas and 44% to

stimulate migration from urban to rural areas (UN, 2009). Some countries (Kenya, Nigeria, Ghana and

Uganda) are also starting to experiment with policies to attract nationals (especially professionals) back in

11 For more details on progresses of SSA countries with respect to these policies, refer to UN (2009).

26

the country (Adepoju, 2008). Population policies, however, do not act in a vacuum. In order to fully reap

their benefits, countries need to invest in women education and need to adopt other broad measures to foster

socio-economic development (e.g., Jeejeebhoy, 1995; Subrahmanian, 2002).

27

Table 2. Policies for employment creation in SSA

Policy area IP tool Botswana Cameroon Cote

d’Ivoire Ethiopia Ghana Kenya Mauritius Nigeria Rwanda Senegal

South

Africa Uganda Zimbabwe

Trade policy

Export

promotion x x x x x x x x x x x

SEZs x x x x x x x x x x x x

FDI attraction

for export x x x x x x x x x x

Standardization

and quality

controls

x x x x x x x x

Selective trade

protection x x x x x x x

Sectoral

policy

Preferential

credit x x x x x x

Competition

regulation x

Public

procurement x x x x x x

Innovation

policy

FDI attraction

for technology

transfer

x x x x x

Incentives for

equipment and

machinery

x x x x

Industrial R&D x x x x x x x x

28

Clusters x x x x x

Employment

policy

Training x x x x x x x x x x x x

SMEs x x x x x x x x x x

Measures for

rural poor and

informal sector

x x x x x x x x x x

Public works x x x x x x x

Employment

services x x x x

Source: Authors’ elaboration based on: Altenburg (2010), Assefa (2008), Bategeka (2012), ILO (2004), Marti and Ssekubuge (2009), Rizzo (2011), Soludo et al. (2004), UNDP

(2005), UNECA (2011b), Zeng (2008), Walther (2006) and national policy documents.

Notes: Measures for rural poor and informal sector include measures for agricultural development, programs to provide rural poor with the conditions to move to other

productive sectors (e.g. education, technical skills, and access to finance) and measures to incentivize formality and improve productivity of the informal sector. Public works

refer to subsidized temporary employment or labour intensive mega-projects financed to create employment (e.g. infrastructure projects). Employment services refer to active

labour market measures and efforts to develop effective labour market mediation, information and careers advice institutions and services, both in the public and private sector.

29

Table 3. Targeted sectors of industrial policy

Country Sectoral policies

Botswana auto, beverages, textiles and clothing

Cameroon textiles and clothing, wood, energy and hydrocarbons, agro-processing, pharmaceuticals, tourism

Cote d’Ivoire agro-processing, construction, civil engineering

Ethiopia leather and leather products, textile and clothing, sugar, metal, dairy and meat, horticulture, agro-processing, construction

Ghana agro-processing, ICTs, metal-based industries

Kenya agro-processing, fertilizers, cement, fish, leather, pulp and paper, metals, plastics, textiles and clothing, footwear, ICTs, electrics

Mauritius ICTs

Nigeria pre-chemicals, machine tools, steel

Rwanda agro-processing, ICTs

Senegal Tourism

South Africa auto and auto components, textiles and clothing, pharmaceuticals, plastics, metals, pulp and paper, furniture, chemicals

Uganda agro-processing, textiles and clothing

Zimbabwe agriculture, raw materials, clothing

Source: Authors’ elaboration based on Altenburg (2010), Marti and Ssekubuge (2009), Soludo et al. (2004), UNECA (2011b), national policy document.

30

Table 4. A review of the literature on policies for productive employment creation

Policy Area Source Countries covered Main conclusions

SEZs

Cling et al. (2005) Madagascar Average wages in the Zone Franche are equivalent to other formal sectors but labour standards are higher than average.

Rojid et al. (2008) Mauritius Mauritius reduced unemployment and raised foreign exchange, yet EPZ bore more costs than

benefits to the economy.

Kingombe and te Velde (2012) SSA

Half of the EPZ employment in SSA is in South Africa. Ghana, Tanzania, and Kenya are the

countries where the share of SEZs employment is the highest. In few countries, SEZs helped to

change export structure and upgrade economy (Mauritius and maybe Kenya).

FDI attraction

Ndikumana and Verick (2008) SSA FDI crowds in private investment but the impact of private investment on FDI is stronger and more

robust than the reverse relation.

Phelps et al. (2009) Ghana, clothing industry Ownership of FDI matters for insertion in GVCs and development. Rapidly emerging Asian MNEs

present challenges for industrial upgrading via FDI.

Standards and

quality controls

UNECA (2008) Ghana, Kenya, Uganda,

Zambia and Ethiopia

Food safety and quality standards, especially from the EU, undermine the participation of small-

scale growers into GVCs. Small-scale farmers need substantial financial support to achieve these

certifications. Kenya is a good example of proactive strategies in this respect.

Goedhuys et al. (2006) Tanzania In the manufacturing sector, ISO certifications positively affect firms’ productivity.

FDI for

technology

transfer

Portelli (2006) Tanzania

Knowledge transfer occurs mainly via backward linkages with local suppliers. The strength of these

linkages depends on factors like the orientation of the MNEs (domestic or export market), its origin,

and varies across industries. Local capabilities are a prerequisite to seize the benefits of FDI.

Managi and Mulenga Bwalya

(2010)

Kenya, Tanzania and

Zimbabwe

Results show evidences in support of intra- and inter-industry productivity spillovers from FDI for

Kenya and Zimbabwe, but not for Tanzania.

Osabutey and Debrah (2012) Ghana and SSA Countries like Ghana have improved their investment climate to increase FDI. However, policy paid

little attention to technology transfer and capacity building. FDI policies are not integrated with

31

education and technology policies, and private-sector development policies.

Elmawazini and Nwankwo

(2012)

Côte d’Ivoire, Kenya,

Madagascar, Senegal, and

South Africa

FDI inflows have had relatively little impact on SSA’s industrial capacity and global competitiveness

and widened the gap with developed economies. Predominance of FDI in extractive instead of

manufacturing industries and weak absorptive capacity reduce MNEs incentives to transfer

knowledge.

Clusters

Oyelaran-Oyeyinka and Lal

(2006)

Ghana, Suame and Kenya,

Kamukunji and Kariobangi

Managers of firms in clusters are more inclined to train workers if ICT facilities are available within

the cluster. Cluster performance can be improved by taking joint actions in the form of technological and human resource development programs.

Kinyanjui (2008) Kenya, Kamukunji

metalwork Cluster

Sanitation, lighting, electricity, and links with learning institutions must be upgraded if the cluster is

to continue to become more productive.

Oluyomi Abiola (2008) Nigeria, Otigba Computer

Village Cluster The cluster benefited both from initiatives of associations (like CAPDAN), the National Information Technology Policy and public procurement of locally assembled computers.

Kiggundu (2008) Uganda, Lake Victoria

Fishing Cluster

With the help of EU buyers, the cluster has made some progress in process upgrades, less in product

development. It has not yet shifted from the preparation and export of whole and semi-processed fish

products to further processed products.

Sawkut (2008) Mauritius, Textile and

Clothing Cluster

The government provided training, export promotion, SMEs support, favourable business

environment. Given that productivity is lower and wages are higher than global competitors,

Mauritius is pointing at the knowledge-intensive segment (high quality products) of the industry.

Gebreeyesus and Mohnen

(2013)

Ethiopia, Mercato footwear

cluster

Business and knowledge interactions are co-occurring in the cluster. Intense competition from

imports and later on within the cluster has led to some form of upgrading and quality improvements.

Mano et al. (2011) Ethiopia, 64 cut flower

farms

Relatively to firms outside the cluster, firms within the cluster attract workers with higher human

capital, have more permanent than seasonal workers and better cope with demand fluctuations due

to information flows.

Sonobe et al. (2012) Ghana, Ethiopia, Tanzania,

Vietnam, China,

Basic management training helps to improve management practices, reduces the incidence of exit,

and is likely to benefit participants enough to justify the cost of providing the training (due to

32

Bangladesh spillovers within the cluster). Training, however, is not enough to spur firms’ innovation because

innovation capabilities are more difficult to teach.

Yoshino (2011) SSA

Cluster-based enterprises are more productive and export-oriented than outside firms. Firms in

survival clusters fail to grow due to lack of differentiation and lack of opportunities for spatial

mobility. In terms of employment, cluster-based firms absorb more permanent workers, while outside

enterprises more (unskilled) apprentices.

Training

Sekwati and Narayana (2011) Botswana

Current VET set up is not responsive to the needs of the people who did not have formal education

and work already in the informal sector. The 2007 informal sector survey provides an idea of the

training needs of informal sector, but much more detailed analysis is needed for a VET reform.

Okunola et al. (2010) Nigeria

The main problems of VET are: limited resources for expansion, exclusion of VET from the main

stream curriculum, lack of guidance services, quantity and quality of teaching resources. A public

perception of VET as low status has partly vanished government efforts to reform VET.

Berthelemy (2006) and

Biavaschi et al. (2013) SSA

Education policies are biased against primary education and VET has low priority with respect

classical curricula. This bears important distributional consequences.

Monk et al. (2008) Ghana

Apprenticeship is by far the most important form of training in urban Ghana. The most important

factor affecting returns to apprenticeship is the level of prior formal education: for currently

employed people, who did apprenticeships but have no formal education, apprenticeship increases

their earnings by 50%, but the return declines as education rise.

SMEs

Wignaraja (2002) Mauritius

SMEs have lower capabilities, are less export-oriented, have less foreign equity, conduct less

training and make less use of external technical assistance than large firms. Policies can improve

their capabilities and competitiveness by incentivizing co-location with large firms into clusters.

Mckenzie (2011) SSA African SMEs are small and heterogeneous, which poses challenges for both experimental and

structural methods of estimating the impact of firm policies. Different procedures should be used on different set of firms (microenterprises, larger firms, subgroups of relatively homogeneous SMEs).

Cho and Honorati (2013) Developing countries Programs promoting self-employment and small scale entrepreneurship can lead to increases in

33

labour market outcomes. Policy mixes (including training, facilitated access to finance, and

counselling) are more effective than single measures and impacts are higher for young and educated

people.

Policies for rural

areas and

informal sector

Akpan (2012) Nigeria

Policies for rural areas are associated with (subsistence) agriculture and have the imprint of short-